Does anyone has an idea what happend to the “Anonymous Remailer”.
Some years ago, there was an active scene of remailers in order to post anonym into the UseNet or send mails without a sender.
As far as I know, there have even been technical solutions to problems like finding out whether someone is writing something based on traffic. I remember that there were even concepts for a kind of mailing list that worked in principle while respecting privacy.
Has this been developed further?
You mention tor. I don’t thing its an accident. tor and the mixmaster has a lot in common in my eyes. Both use a model of 3 levels at the default (entry, mid and exit), both use a similiar principle to avoid tracking by the time when informations floud throught the system.
You can have both. Anonym communication and social media with posting pictures of your food. But with things like mixmasters, it’s remains your choice which level of privacy you need.
P.S.: Maybe, I should ask on reddit or in a technical forum about it?
I suppose so, but the overall picture may be about the same. Not just technical but also sociological changes have made this stuff less attractive. And even if your mixmaster or tor traffic is really anonymized (which is dubious), the fact you are using such services at all probably flags you for attention. If you just want to exchange email with your friend when you are both on the down low, you might be best off just both enrolling gmail accounts. There are so many gmail users that being one won’t attract attention, and with both people on the same service, all the traffic stays inside the google network and might be harder for outside agencies to connect with individual users (maybe foolish optimism). Plus, a lot of attention in the crypto nerd world shifted over to things like bitcoin.
I have heard there was a recent development in single server PIR (private information retrieval). I haven’t yet tried to understand how it works, since it uses fancy cryptography (homomorphic encryption). In simplest form, let’s say there is a database with a billion records and you want record# 123456. PIR means being able to retrieve that specific record without the server learning which record it is that you want. There are known ways to do that by spreading the query between multiple, non-cooperating servers, but it was long believed that the only way to do it with a single server was for the client to download the entire database. This recent discovery apparently gives a way to do it with just one server, at some intermediate computational and network cost. That is of potential interest for this sort of application.
Thank you for your long answer. Even if it doesn’t contain the answer I looking for.
I think so, too. But in my imagination, one who uses tor or a remailer is just a short flash on the radar. If it doesn’t follow more, they would not investigate further. Since 90% or so of all tor users doesn’t do anything bad. Even agencies doesn’t like wasting of time and resources.
With Google, I think, it will nearly 100% sure that it is tracked somehow. If both sides have no problem with this, its fine.
Never considered this angle. Thanks for this.
I remember, I have read, over 5 years ago, a PDF-File from a German university about this. It was a so called “blinded read”-method.
I have a vagly idea how this can work but I lack the mathematical knowleade to explain it further.